融合隶属度函数的自适应惯性权重模式的粒子群优化算法

粒子群算法(Particle Swarm Optimization,PSO)的性能极大地依赖于其惯性权重参数的选择策略。当在一次迭代中更新粒子速度时,PSO忽略了粒子间的差异,在所有粒子上应用了相同的惯性权重。针对这一问题,提出一种自适应惯性权重的粒子群算法PSO-AIWA,有效合理地均衡PSO的全局搜索和局部搜索能力。根据当前粒子与全局最优粒子间的差异,算法可以通过基于粒子间距的隶属度函数动态调整粒子的惯性权重,使得每次迭代中,粒子可以根据当前状态在每个维度上的搜索空间内选择合适的惯性权重进行状态更新。在6种基准函数下进行了算法的性能测试,结果表明,与随机式惯性权重PSO算法与线性递减惯性权重PSO-LDIW算法相比,该算法可以获得更好的粒子分布和收敛性。

PARTICLE SWARM OPTIMIZATION BASED ON ADAPTIVE INERTIA WEIGHT MODEL WITH MEMBERSHIP FUNCTION

The performance of particle swarm optimization(PSO)depends greatly on the selection strategy of its inertia weight parameters.When updating the particle s velocity in one iteration,PSO neglects the difference among particles and employs the same inertia weight for all particles.To deal with this problem,we propose PSO based on adaptive inertia weight adjusting(PSO-AIWA)to rationally balance the global searching and local searching abilities for PSO.According to the difference between the current particle and the global optimal particle,the algorithm could dynamically adjust the inertia weight of the particle through the membership function based on the particle spacing,so that the particle could choose the appropriate inertia weight to update the state on the basis of the current state in the search space of each dimension in each iteration.The performance of the algorithm was tested on 6 benchmark functions.The results show that compared with the stochastic inertia weight PSO and the linear decreasing inertia weight PSO-LDIW,PSO-AIWA can obtain better particle distribution and convergence.